Good Management Practice Is Correlated With Good Performance of Community-Engaged Primary Health Care Facilities in Peru

This study aims to contribute to a better understanding of regional management practices that could support primary health care service performance in the context of the innovative community engagement model implemented through the unique program with embedded mechanisms for accountability.


INTRODUCTION
2][3] The Alma Ata Declaration on Primary Health Care of 1978 was signed by 134 countries 4 ; 120 of them reconfirmed their commitment with the 2018 Declaration of Astana. 5[8][9][10][11] In 2024, the World Health Organization (WHO) Alliance for Health Policy and Systems Research identified 5 focus areas where health policy and systems research can make a unique contribution to improving health and health equity, the first being the reform of health systems and services to achieve UHC through a PHC approach. 12he urgency of equitable health systems reform was made apparent in the major failings of PHC in many low-and middle-income countries during the COVID-19 pandemic: 4.5 billion people were not covered by essential health services in 2021 and "less than a third of countries have improved health service coverage" over the past 2 decades, with disproportionate consequences for the poor. 3he 2020 WHO operational framework for PHC provided 4 strategic and 10 operational levers to guide the substantial reforms that are required to meet UHC goals, the 4 main levers being political commitment and leadership, governance and policy, funding and allocation of resources, and engagement of communities and other stakeholders. 2This and other conceptual frameworks, such as the Primary Health Care Performance Initiative 6 and Essential Public Health Functions, 13,14 help countries identify what works for better delivery of quality populationbased PHC services and where more innovation and research are needed.
Governance and leadership are critical strategic levers for PHC but have had relatively little research published on them compared to other domains. 15,16lthough governance and management comprised 1 of WHO's 6 "Building Blocks," there was little consensus on key concepts and measures of it. 17efinitions of PHC governance now include quality management infrastructure that has varied aspects and several operational levels; it can refer to the way services are managed in health facilities or at higher levels, such as district or subnational management units. 6PHC systems with optimal management practices are part of the structural innovations that are "more likely to improve service quality substantially and at scale than are micro-level efforts." 18n general, management practices have been categorized as leadership, planning, monitoring, supervision, and evaluation. 19Evaluation detects differences between the actual and desired situation based on planning.Monitoring and supervision are mechanisms to provide information for evaluation and subsequent replanning.
The Primary Health Care Performance Initiative suggested roles for PHC management as qualified management and leadership, information systems to support and improve quality, budgets and financing systems to meet essential costs, management capability and leadership, and supportive supervision for continual improvement of knowledge and skills. 6he WHO PHC operational framework proposes a set of governance roles at the subnational level (Box 1). 2 Good practices for management of health policy implementation cannot be generalized and need to be tailored to each country. 20When policies lack clarity on management responsibilities, scant priority is assigned for effective control of PHC from national and subnational levels.2][23][24] Traditional public PHC services in Peru are managed from the top down for all goods and services.Local health facilities have no responsibility for planning, financial, or human resource management and have no accountability or community engagement mechanisms.The focus of supervisory action is on technical protocols and information reporting that is exclusively health service related.
In a study of Peruvian government PHC services, health facility medical officers were asked to identify the most common PHC management problems. 25Their responses reflected the most generally identified PHC service delivery bottlenecks, principally focused on human resource shortages, lack of supplies and medicines, poor infrastructure, and budget deficits but not on problems of management per se.A perception of responsibilities in a multilevel working system of clear roles, accountabilities, and principal-agent relationships was lacking in the responses provided.
Community engagement is another domain that has again been brought to the fore by WHO and firmly established as a strategic lever in the operational framework for PHC-both as part of local governance in terms of community voice and oversight for greater social accountability and as a component of population health management, reaching outside the health facility for proactive community-based education and monitoring. 2,14 systematic review of 111 studies was conducted in 2019 on what works and does not work to strengthen PHC systems in low-and middleincome countries.Five areas of intervention were found most likely to be effective for improving the Governance and leadership are critical strategic levers for PHC but have had relatively little research published on them compared to other domains.
desired outcomes of quality, coverage, efficiency, responsiveness, and equity of PHC service delivery: (1) placing greater emphasis on use of nonphysician health workers, (2) integrating noncommunicable disease prevention and control into the basic package of care, (3) modernizing PHC information systems, (4) building managerial capacity, and (5) institutionalizing community engagement. 26Assessment of managerial capacity was found lacking in nearly all the studies reviewed; nevertheless, it was highlighted as "one of the interventions most likely to improve performance outcomes," being necessary to operationalize all the other facility-level interventions.Furthermore, the evidence review showed that improvement of all desired outcomes of PHC services was strongly associated with community engagement.
The current study uses a systems-thinking approach to understand an important health system reform in Peru that strengthens PHC.Specifically, this study aims to contribute to a better understanding of regional management practices that could support PHC service performance in the context of the innovative community engagement model implemented through the unique national Shared Administration (SA) Program that has various embedded mechanisms for accountability.Established 30 years ago, community involvement in PHC services in Peru was institutionalized in 1994 through the formation of legally recognized registered community-based nonprofit organizations called Comunidad Local de Administración de Salud (Local Community for Health Administration, CLAS) under the SA program. 27The program was so popular that by 2002, it had scaled up, based on demand, to encompass nearly one-third of all public PHC facilities in Peru.
A 2007 survey, unreported until now, of Dirección Regional de Salud (regional health directorates, DIRESAs) provided the opportunity to assess their management practices in relation to CLAS-comanaged PHC facilities at a time when there was relatively strong national support for the SA program.9][30] Furthermore, a national law on comanagement and citizen participation in PHC services was passed in 2007 by the Peruvian Congress to provide legal stability to the SA program and CLAS facilities. 31However, since that time, opposition from interest groups, faulty decentralization efforts, an apparently competing model of health service organization, and budget cuts contributed to the eroding of Ministry of Health (MOH) support to CLAS comanagement of PHC facilities.More detail is provided in the Discussion section.
We posit that the 2007 survey remains relevant for assessment for retrospective learning on what works.In the post-pandemic era, a renewed realization that PHC should be prioritized provides the impetus for the consideration of CLAS facilities as a solution to reach UHC.Thus, it is relevant to the future expansion of CLAS facilities to assess factors that helped to make them work well.Furthermore, the 2007 survey results could provide guidance to other countries that are starting to design their own models of PHC with community engagement to better reach UHC by 2030.

COMMUNITY COMANAGEMENT MODEL THROUGH THE SHARED ADMINISTRATION PROGRAM
The most important policy and structural change of the MOH for PHC in Peru in the last 3 decades was the innovative design and institutionalization of a community comanagement model through the SA Program.The program was established in BOX 1. Subnational-Level Governance Roles in the World Health Organization Operational Framework for Primary Health Care Reform and align the integrated primary health care (PHC)-oriented governance mechanism and planning processes at the subnational level to respond to its 3 components (integrated health services including public health functions, community participation, and multisectoral collaboration).
Create community-based multistakeholder forums for collective accountability and action on health and health-related issues.
Create an organizational culture that supports monitoring and evaluation through knowledge-sharing, open feedback, and a demand for data in decision-making processes.
Strengthen PHC-oriented management protocols that encourage provider report cards, patient satisfaction surveys, patient-reported outcomes, and balanced scorecards.Support public, private, and community actors to develop competencies for engaging across the PHC components.
Source: World Health Organization. 2 The current study uses a systemsthinking approach to understand an important health system reform in Peru that strengthens PHC.
1994 by Supreme Decree (a legislative instrument just below the level of a law) that legally recognized CLAS.Beginning with 16 PHC facilities collaboratively administered by CLAS in 1994, by 2006, as many as 764 CLAS had been formed, comanaging 2,155 (31%) PHC facilities of a total of about 6,700, benefiting 7 million inhabitants comprising 80% of the poverty population in Peru. 27In 2007, the SA program and the comanagement of PHC facilities by CLAS were formalized into national law with regulations to the law approved in 2008. 31,32he comanagement model for PHC has 7 innovations that incorporate accountability mechanisms, none of which are present in traditional public administration of the health sector. the community is involved in actions that integrate essential functions of public health with PHC services and legal accountability mechanisms for those actions.
Previous studies comparing CLAS facilities to non-CLAS facilities have shown better performance of CLAS facilities on indicators of quality, coverage, equity, efficiency, and impact. 28-30 ,34-38 For example, patient waiting times of less than 15 minutes and more than 30 minutes were found in 75% and 6% of CLAS facilities, respectively, and in 43% and 29% of non-CLAS facilities, respectively. 28Health service coverage was measured in a study of all 675 PHC facilities in 3 health regions (Cusco, Huánuco, and La Libertad), showing the average number of health visits per child younger than age 5 years per year as 2.32 and 2.94 in rural and urban CLAS facilities, respectively, while the averages were 1.32 and 1.73 in rural and urban non-CLAS facilities. 30he current study aimed to determine if good regional management practices were correlated with better performance of CLAS facilities and identify good management practices that may be linked with better PHC service performance in CLAS facilities.

METHODS
A secondary analysis was conducted of a survey of 23 DIRESAs implemented in 2007 by the MOH of Peru.The Introduction and Discussion sections examine the relevance of reporting on the 2007 survey.The analysis aimed to assess the correlation between the use of good regional management practices and the level of service performance in CLAS comanaged facilities.Management actions were assessed through multiple-choice survey questions.More than 1 response could be marked for each question, which was classified by the authors according to the management function framework of Primary Health Care Performance Initiative 6 : leadership of the SA management team; information systems (planning, goals, monitoring); financial control; transfer of health facility management capacity and leadership skills; and supervision.A panel of experts on the SA program and CLAS classified responses as either "good practice" or "less effective practice."We tabulated the number of good management practices reported by each DIRESA and divided that sum by the total number of possible good practices to obtain the percentage of good management practices used by each DIRESA.
As the primary outcome variable, the level of performance on delivery of PHC services was defined as the fulfillment of goals established in the local health plan of each CLAS.These goals were committed to in the comanagement agreement (contract) between each CLAS facility and its respective DIRESA.We used the inverse of the DIRESA response to the survey question, "What proportion of existing CLAS in the region have demonstrated limitations or deficiencies in meeting contracted goals?"Goals referred to the planned achievement of health coverage indicators (e.g., population-based coverage of child immunization, well child visits, and prenatal care) and health service utilization indicators (e.g., numbers of medical consultations and follow-ups, emergency visits, and dental consultations).
Pearson's correlation coefficient was used to statistically describe the linear relationship between the percentage of good management practices used by each DIRESA and their respective average proportion of CLAS facilities with good service performance (goal achievement).
To identify specific practices associated with good performance, the DIRESAs were divided into 2 analysis groups: Group 1 with "higher-performing PHC services" and Group 2 with "lower-performing PHC services" (Table 1).
We compared Groups 1 and 2 using chi-square to verify the differences between them in the percentage of good management practices used and in the percentage of contracted goals achieved.
To rule out external confounders that might explain the differences between the 2 study groups of DIRESAs, the groups were compared on overall regional health system capability and equity considerations by assessing the mean infant mortality rates in each of the 2 study groups during the 5-year time periods before 1996, 2007-2008, and 2014 using national demographic and health survey results.
Finally, good management practices were identified that were reported by the majority of both Groups 1 and 2 or reported more frequently in Group 1 versus Group 2.

RESULTS
Twelve of 23 DIRESAs responded to the survey, reporting on 1,381 CLAS facilities in their regions, which comprised 64% of all 2,155 CLAS facilities nationally.On average, 65.4% of 1,381 CLAS facilities in 12 DIRESAs met the contracted goals in their respective comanagement agreement.
Thirty-two of the 52 possible management actions in the survey response options were classified as good management practice and 20 as less effective.An average of 32% of the 32 possible good practices were implemented across all DIRESAs surveyed.
A secondary analysis was conducted of a survey of 23 DIRESAs implemented in 2007 by the MOH of Peru.
The Figure shows a significant correlation between PHC services performance and the use of good regional management practices.Pearson's correlation coefficient was r¼.7266, 12 df, P<.01.
Dividing study DIRESAs into 2 analysis groups based on meeting performance goals, DIRESAs in Group 1 reported an average of 91.1% (range 89% to 100%) among 718 CLAS facilities that met their performance goals, while Group 2 reported an average of 37.6% (range 20% to 74%) among 663 CLAS facilities.The difference between the 2 groups is significant at P<.01 (Table 1).
Assessing good management practices by study group, DIRESA Group 1 used an average of 40.6% of good practices considered, significantly different from DIRESA Group 2, which used an average of 24.0% (Table 2).

Good Management Practices in the Majority of Regions Studied
The majority of DIRESAs studied used the following 4 of 32 good management practices: (1) in 11 of 12 regions, CLAS submitted monthly financial reports with supporting documents to the DIRESA; (2) 9 of 12 DIRESAs provided technical advice to their respective CLAS associations on the formulation of their local health plan; (3) the TTSA jointly coordinated and/or conducted supervisory field visits to CLAS facilities along with supervisors of other vertical health programs; and (4) the use of a preestablished supervision guide was the case in the majority of regions.

Good Management Practices in Regions With Better Primary Health Care Service Performance
Eleven of 32 good practices were reported more frequently in DIRESAs of Group 1 than Group 2. "More frequently" was defined as a difference of 33% or more (2 or more of 6 DIRESAs) in Group 1 reporting a certain practice compared to the number of DIRESAs reporting it in Group 2. The good management practices reported more frequently by Group 1 included 2 aspects of the TTSA formation; 1 aspect of the TTSA work planning; 1 aspect of CLAS supervision; 2 aspects of financial control; 3 aspects of technical assistance provided to CLAS; 1 aspect of human resources management; and 1 aspect of health promotion planning (Box 2).
Detailed results of all survey questions by study group are shown in Supplement 2 (English) and Supplement 3 (Spanish).DIRESA Group 1 used an average of 40.6% of good management practices considered compared to DIRESA Group 2 that used an average of 24.0%.

DISCUSSION
PHC system design should include optimal management practices to ensure quality, efficiency, and effectiveness.These are part of the structural innovations that are identified as crucial for quality services, "more likely to improve service quality substantially and at scale than are micro-level efforts." 18revious evidence has shown the positive effects of community involvement on PHC service delivery performance or quality, 26 but few prior studies have assessed how good subnational management practices could potentiate the effect of community involvement on service improvements.
The finding that regional-level infant mortality trends are similar in DIRESA Groups 1 and 2 supports speculation that differences in the main study variables were not due to mean differences in regional capabilities, resource availability, or population characteristics between the 2 study groups.
Despite the existence of evidence of the superior impact of the CLAS comanagement model on PHC service delivery as compared to standard government PHC services (non-CLAS), health sector support for CLAS has waned over time and no longer enjoys the same level of political mandate or budgets for full implementation of the CLAS law for reasons that are important to recognize.Flagging support to CLAS may be related to the following 4 factors.
First, public sector health worker unions have threatened and conducted national strikes to get health personnel contracts changed to appointments to permanent public sector payroll, referred to as nombramiento to government service.Permanent appointments have 3 main effects on CLAS facilities, especially those in rural locations: (1) reduced PHC productivity at greater cost (lower efficiency) because the permanently appointed staff work 6 instead of 8 hours per day without proportional salary adjustment; (2) loss of PHC staff because permanently appointed staff can request transfer to urban hospitals or to publicly paid residency training, debilitating quality in public PHC services; and (3) reduced effectiveness of management through less accountability of permanently appointed staff. 42Medical professionals rejected the idea  of being contracted by the community through CLAS facilities. 43They also unfairly blamed CLAS facilities for paying lower salaries, even while inadequate funds were provided for more needed staff to meet the ever-increasing demand for the better-quality services that CLAS facilities were providing.Supervision of public PHC services (mainly non-CLAS facilities) in 2018 by Controlería de la República (the national oversight agency) revealed physician productivity in number of consultations per hour to be 1.22, while the expected number is 3.72. 44 majority of health workers now are appointed staff with permanent public payroll status.Solutions to this issue could include providing adequate incentives (e.g., standardizing PHC worker salary levels) that would make working conditions more attractive in CLAS facilities and could constitute a superior alternative to public payroll contracts to achieve higher quality and better health outcomes for Peru's vulnerable population that depends on public services.
A second factor that reduced support to CLAS facilities was created by the 2002 law on decentralization and 2003 law on regional governments, which were construed by subnational governments as conferring to them discretion on whether to implement national health policy.Subnational regions with all or nearly all PHC services incorporated into the SA program continue to support their CLAS facilities.Six of 25 regions have 60% to 100% of their PHC facilities incorporated into CLAS: Arequipa (100%), Tacna (100%), Madre de Dios (100%), Moquegua (91.7%),Huánuco (68.2%), and Tumbes (61.9%). 27Others with fewer CLAS facilities have tended to have a lower level of interest and compliance with the program.The solution to this problem would be to have a strong national mandate to implement the CLAS law and continual capacity-building of regional governments on CLAS law and regulations.
A third factor has been the introduction of a broad and well-financed plan to create Redes Integradas de Salud (integrated health networks, RIS) with cooperating networks of PHC services in 3 public health services subsystems.One is the least-financed MOH-Regional Government public system that covers 68% of the population, and the others are the Social Security subsystem (based on obligatory formal-worker payroll deductions) and BOX 2. Good Management Practices That Were More Frequent a in Group 1 Than in Group 2 Leadership (of the subnational management team) The Shared Administration (SA) program coordinator was a regional health directorate (DIRESA) official specifically designated for that purpose.
The coordinator of the SA program was designated by Directorial Resolution.
Information systems (planning, goal setting, monitoring) Health education and information goals were incorporated into the local health plan.

Financial control
Expenses incurred by each Comunidad Local de Administración de Salud (CLAS) were monitored through periodic visits to primary health care facilities to corroborate the existence of supporting documents.
Balance sheets of CLAS were audited by an external certified public accountant contracted by DIRESA.

Transfer of facility management capability and leadership skills
Trainings for staff and CLAS community members were conducted.
Instructional and informative documents were distributed.DIRESA personnel were assigned to carry out procedures for the benefit of the CLAS.

Supervision
The Technical Team for the Shared Administration Program (TTSA) was clear about the actions that must be implemented for acceptable development of CLAS.Supervision visits to CLAS facilities were performed exclusively by TTSA members.

Other resources provided to CLAS
Newly graduated health professionals were contracted short term.
a More frequent defined as reported by 2 or more DIRESAs in Group 1 over the number reported in Group 2.
the Armed Forces and Police subsystem.In the short-to-medium term, RIS creation is being applied to only the MOH-Regional Government PHC system that is managed logistically by intermediate-level health network management-budget executing units, which will continue to provide logistics management support to RIS.An RIS does not have its own logistics management unit, a mechanism for community comanagement, nor a planned system for subnational management support or oversight.The preexistence of CLAS within the jurisdiction of a new RIS has generated questions about whether CLAS can coexist with RIS, even though it is clear to many that RIS and CLAS are complementary and that, furthermore, an RIS might operate better as a CLAS.
Fourth, funding of public PHC services, including CLAS facilities, was consistently provided by the public Seguro Integral de Salud (integrated health insurance program, SIS) beginning in about 2002.However, in 2016, SIS began to instruct that funds to CLAS facilities should primarily cover variable health care costs focused on medicines, disallowing CLAS facilities to use per capita funding to hire staff.Also, SIS increasingly restricted use of funds by CLAS facilities for fixed costs, such as utilities, security, laundry and cleaning services, ambulance and building maintenance, equipment, and others.Since 2022, SIS payments directly to CLAS facilities have mostly stopped and now go to intermediate-level budget executing units, which are frequently inefficient in spending large budgets to supply PHC facilities in their jurisdiction in a timely manner for patient needs.As a result, now CLAS facilities, as well as non-CLAS facilities, frequently lack medicines, medical supplies and materials, laboratory and X-ray equipment, goods, and services.
There are advocacy efforts now to reverse these decisions.Stopping the transfer of funds to CLAS has been judged by MOH lawyers to contravene CLAS law that requires that public funds be transferred to CLAS for human resources, medicines, supplies, and other local expenses to ensure quality of care to the public. 31urthermore, public health insurance reimbursements and reforms to integrate health subsectors focus primarily on clinical services and have few provisions for community health, empowerment, or social and behavior change, which now affect CLAS as well as non-CLAS facilities.The latter has long been subject to a biomedical approach to PHC provision.
Nevertheless, CLAS facilities remain an important presence in the 6 (of 25) regions in which 60% to 100% of their respective PHC services are comanaged by CLAS.That the CLAS comanagement model of PHC has survived opposition is a credit to its resiliency related to evidence-based results, [28][29][30] legal stability conferred by law, 31 and advocacy by satisfied clients who attend CLAS-run PHC services, health staff who work in CLAS facilities, and by program champions.The 4 factors previously discussed could be reversed through renewed commitment to and compliance with the CLAS law by the MOH and regional governments.
Structural aspects of community involvement and associated accountability enshrined in CLAS law provide an underlying explanation for the study findings.Empowered community that demands accountability is a key structural innovation for high-quality services that was identified in a literature review using the Lancet Global Health Commission framework on high quality health systems. 7

Limitations
Potential limitations to this study include that only slightly over half of DIRESAs responded to the survey.However, results do represent the situation for nearly two-thirds of all PHC facilities in the SA program comanaged by CLAS.The survey was limited in the level of detail on management actions, not all management actions were measured, nor were all factors influencing goal achievement or the relationship between management and goal achievement.The age of the survey is not a limiting factor in answering the study question because the survey reflects a time when political support for and compliance with the SA program was stronger-these being basic prerequisites for successful PHC services with community involvement-and allow an assessment of how things worked under those minimally supportive conditions.

KEY GOOD MANAGEMENT PRACTICES
Given the significant correlation found between good subnational management practices and PHC services performance among CLAS facilities, the following 7 good management practices can and should be strengthened in community-engaged PHC systems.Leadership: Clarity about responsibility is key to the proper functioning of a program. 19A formally recognized leader has the authority to hold team members accountable.Formalization of leadership with designation by documentary resolution is a regulatory requirement for assuming responsibilities in the Peruvian public sector.Two best practices in this study that were more frequent CLAS facilities remain an important presence in 6 of 25 regions in which 60% to 100% of their respective PHC services are comanaged by CLAS.
in Group 1 were the TTSA coordinator being expressly assigned as such and being formally designated with official documentation.Group 2 was more likely to have a TTSA coordinator who had other responsibilities, such as being the overall director of health services or coordinator of vertical health programs, with no formal designation as TTSA coordinator.Lack of formalized leadership designation may contribute to a lack of program commitment and compliance and should be included in an expanded definition of leadership in public governance.Information systems (planning, goals, and monitoring): It is essential to recognize the progression of planning, implementation, and evaluation. 45At the time of the 2007 survey, there was no requirement that TTSAs have an annual work plan to identify and schedule management activities, which helps to explain why such plans were infrequently reported in the survey.However, TTSAs of DIRESAs in Group 1 were more often reported to be "very clear in the actions they needed to perform to ensure acceptable outcomes of the SA program" as per the survey response item.In contrast, Group 2 TTSAs made plans only periodically depending on the prioritization of needs or activities.After the survey was conducted, the central MOH formalized a requirement for TTSAs to develop a work plan.Health promotion planning.Health promotion is one of the main objectives of CLAS as stated in regulations to the law. 32When health promotion activities are included in the local health plan to provide community health education and information and to address social determinants of health, its financing and implementation are more guaranteed through the comanagement contract. 45,46Specific local health promotion strategies should be designed and developed with CLAS community member participation and included in the local health plan.If health promotion is left to the discretion of each PHC service without prior inclusion in the local health plan, compliance is not assured.Financial control.Financial control requires high standards and depends on an organization's accounting and financial reporting. 47CLAS members and managers collaboratively manage the state's financial resources, keep accounting books, report balances monthly to the DIRESA with supporting documents, and are lawfully responsible.Thus, transparency and financial control are more ensured.Monthly reporting by all CLAS was confirmed by 11 of 12 DIRESAs.The TTSA's periodic visits to CLAS facilities to review supporting documents were more frequent in Group 1 than in Group 2. As for audits, DIRESA's outsourcing to a certified public accountant provided greater objectivity and transparency.Auditing by the DIRESA's Finance Office is also considered good practice, though it was not widely reported in the survey.It is not recommended that the audit be carried out by a certified public accountant who is part of the TTSA due to conflict of interest, nor by the DIRESA's Internal Control Office (except in extraordinary cases).Financial support for the TTSA.Neither Group 1 nor Group 2 DIRESAs in this study had a subnational program budget available to maintain the TTSA operational costs.Eleven of 12 DIRESAs received direct financial support from the central MOH office in charge of the SA program to fund TTSAs.This was a less effective management practice for its potential irregularity.When central MOH budget cuts for TTSA were instituted, some but not all DIRESAs took responsibility for these costs.Transfer of health facility management capability and leadership skills.Technical support is essential for proper PHC facility functioning, whether CLAS or non-CLAS.Types of technical assistance that distinguished Group 1 from Group 2 were training provided to CLAS members and PHC staff and distribution of instructional and informative documents.Supervision.Supervision should always have a prior plan with objectives, identification of operational and administrative activities and information to be supervised. 48Supervision of non-CLAS PHC services is oriented to the evaluation of clinical competencies and service-related information.CLAS comanaged facilities also need supervision by vertical health program managers because clinical aspects (clinical protocols, stock of drugs and supplies, health information system, and others) are the same for both CLAS and non-CLAS facilities.In contrast, TTSAs supervise and support compliance with legal, administrative, and accounting obligations of each CLAS facility, supervise the CLAS manager as executor of CLAS decisions, and monitor progress toward achievement of the local health plan.Thus, TTSA needs its own supervision guide.
The aspect of supervision that most distinguished Group 1 was that CLAS facility supervision was performed exclusively by TTSA members.Logically, supervision is most effective when supervisors are highly knowledgeable and experienced in the regulations and processes of the program to be supervised.In Group 2, supervision of CLAS by intermediate-level health network managementbudget executing units was reported more frequently.Though not considered a favorable practice, this could be a good practice in the future if these units receive proper training and experience with the SA program and CLAS law and regulations.

CONCLUSIONS
The study findings suggest that national healthrelated protocols, strategies, and standards may not be sufficient to achieve goals for better PHC service performance.Rather, national PHC strategies should ideally incorporate well-designed plans for operational management from the subnational level and for community-engaged PHC that is designed for accountability.Attention at the subnational level to critical aspects of PHC management (leadership of the management team, management planning, financial management, technical assistance, supervision, and local planning for health promotion and equitable access) could be fundamental for achieving sustainable improvements in health outcomes.The what and how of each management component depends on the design of the PHC-community model and must be continually reevaluated and calibrated.
Public-sector PHC services have proven deficiencies in quality, coverage, equity, efficiency, and impact, due largely to inadequate budgets but also to weak and inefficient accountability provisions, particularly in human resources management with public payroll staff, as well as extremely strict rules for procurement in public sector administration law.The PHC model of the SA program with community comanagement through CLAS is unique in the accountability embedded in its modernized contracting, procurement, and financing arrangements, allowing local decisions at the facility level that are monitored by the community.In turn, these processes are likely to be strengthened by good management from both national and subnational health offices.
Study results support recommendations for Peru to update the CLAS law to strengthen mechanisms for SA management from subnational levels; sustainably strengthen the leadership and financing of TTSA; specify TTSA roles and management instruments for planning, training, and supervision; maintain and reinstate enforcement of CLAS legislation that embeds accountability in local contracting and purchasing under private law; and resume political support for expansion of the SA program and CLAS model to improve coverage, quality, equity, and efficiency of PHC with community participation to reach goals of universal health coverage and the Sustainable Development Goals.
Study findings could help to encourage and guide other countries as they begin to redesign PHC with institutionalized community engagement that includes accountability mechanisms and intentionally align subnational management guidelines with them.It is recommended that PHC management and community engagement mechanisms be continuously evaluated to allow for effective ongoing improvements.
Further research is warranted to validate findings in similar studies with larger sample sizes, as well as studies that compare accountable communityengaged PHC with traditionally administered public PHC services.

FIGURE.
FIGURE.Correlation a Between Percentage of CLAS Comanaged Primary Health Care Facilities That Completed Their Contracted Goals and Percentage of Good Management Practices Used by Each DIRESA

TABLE 2 .
Percentage of Good Management Practices Used to Support CLAS Facilities, by DIRESA and Study Group, Peru Abbreviations: CLAS, Comunidad Local de Administración de Salud; DIRESA, regional health directorate.a DIRESAs with a higher level of compliance with the goals contracted by CLAS.b DIRESAs with a lower level of compliance with the goals contracted by CLAS.c Chi-square difference in study groups with Yates correction: X2¼(1,N¼384)¼11.45,P<.01.

TABLE 1 .
Percentage of CLAS PHC Facilities that Completed Contracted Goals, by DIRESA and Study Group, Peru Comunidad Local de Administración de Salud; DIRESA, regional health directorate; PHC, primary health care.a Completion of goals in comanagement contracts, based on local health plans.b Chi-square difference in the 2 study groups: X2¼(1,N¼1381)¼452.82,P<.01.